Food Chemistry 80 (2003) 119–123

www.elsevier.com/locate/foodchem

Analytical, Nutritional and Clinical Methods Section

Determination of heavy metals in fish samples of the middle Black

Sea (Turkey) by graphite furnace atomic absorption spectrometry
Mustafa Tüzen*
Gaziosmanpaşa University, Faculty of Science and Arts, Chemistry Department, 60250 Tokat, Turkey

Received 5 February 2002; received in revised form 12 June 2002; accepted 12 June 2002

Abstract
The concentrations of heavy metals (Pb, Cd, Fe, Cu, Mn and Zn) in fish samples were determined using graphite furnace atomic
absorption spectrometry after dry ashing and wet ashing methods. Different matrix modifiers were used for the stabilization of the
analyte. Good accuracy was assured by the analysis of biological reference materials. Recoveries were quantitative for all elements
studied (595%). The relative standard deviations were less than 7% for all elements.
# 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Heavy metals; Digestion; Fish; Graphite furnace atomic absorption spectrometry

1. Introduction Decomposition of solid samples is an important step in

The heavy metal pollution of the marine environment
has long been recognized as a serious environmental
concern (Balkas, Tugrel, & Salhogln, 1982; Tarıq, Jaf-
far, & Moazzam, 1991). In the sea, pollutants are
potentially accumulated in marine organisms and sedi-
ments, and subsequently transferred to man through the
food chain (Giordano et al., 1991). For these reasons, it is
important to determine the chemical quality of the marine
organisms, particularly the contents of heavy metals, in
order to evaluate the possible risk, to human health, of fish
consumption (Cid, Boia, Pombo, & Rebelo, 2001). Metals
such as iron, copper, zinc and manganese, are essential
metals since they play an important role in biological sys-
tems, whereas mercury, lead and cadmium are non-essen-
tial metals, as they are toxic, even in traces (Schroeder,
1973; Somer, 1974). The essential metals can also produce
toxic effects when the metal intake is excessively elevated.
Levels of heavy metals in fish have been widely reported
(Pujin, Djukic, Maletin, Obradovic, & Kostic, 1990;
Sharif, Mustafa, Mirza, & Safiullah, 1991; Tarıq, Jaffar, &
Ashraf, 1991; Ubillus, Alegria, Barbera, Farre, & Lagerda,
2000; Voegborlo, El-Methnani, & Abedin, 1999).
* Tel.: +90-356-2521582; fax: +90-356-2521585.
E-mail address: mtuzen@gop.edu.tr, drtuzen@hotmail.com
(M. Tüzen).
0308-8146/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0308-8146(02)00264-9
combined analytical methods. In most cases, when using
highly sensitive measuring methods, such as flame AAS,
graphite furnace AAS, ICP–OES, ICP–MS or inverse
voltammetry, the sample is measured in an aqueous solu-
tion (Knapp, 1991). Combined analytical methods are
favoured for multi element analysis of environmental and
biological samples at very high speed. Sequential and
simultaneous determinations of the elements can be
made using the above analytical techniques. Determi-
nation of heavy metals (Pb and Cd) in fish has been
performed with electrothermal AAS (detection limits of
Pb: 6.4 ng/ml and Cd: 0.14 ng/ml) (Sures, Taraschewski,
& Haug, 1995). The concentrations of Zn, Cd, Cu and
Pb in fish have been determined by using ICP–AES and
FAAS (Zhuang, Wang, Yang, Zhu, & Yang, 1995).
Heavy metal concentrations have been determined in
fish and biological samples using flame AAS with
detection limits (as mg/l) Cd: 0.6, Cu: 3, Fe: 5, Mn: 3,
Pb: 20 and Zn: 2 (Karadede & Ünlü, 2000). Levels of
heavy metals in biological samples have been deter-
mined by flame AAS with detection limits (as mg/g) Cd:
0.03, Cu: 0.03, Zn: 0.07, Fe: 0.04 and Mn: 0.01 (Kress,
Hornung, & Herut, 1998). Trace metals have been
determined in fish samples by electrothermal AAS with
detection limits (as mg/l) Cd: 0.0788, Cu: 0.344, Ni:
1.825, Pb: 0.968 and Zn: 0.0127 (mg/ml by FAAS) (Cid
et al., 2001). Detection limits of molybdenum have been
120 M. Tüzen / Food Chemistry 80 (2003) 119–123

reported in geological samples and seawater, using com-
bined analytical methods, as 0.75 mg/l for ICP–AES, 0.08
mg/g for ICP–MS, 0.18 mg/ml for FAAS and 3 ng/l for
GFAAS, respectively (Santos, Korn, & Ferreira, 2001).
Graphite furnace atomic absorption spectrometry
(GFAAS) is one of the suitable methods for the deter-
mination of trace metals in food and biological samples
because of its speed, minimum need for sample pre-
paration, the possibility of automation, good sensitivity
and low detection limit (Acar, 2001; Acar, Kılıç, &
Türker, 2000; Doner & Akman, 2000; Blust, Vander der
Linden, Verheyen, & Decleir, 1988; Huang et al., 2000;
Lynch & Littlejohn, 1989). However, the determination
of heavy metals in fish samples by GFAAS is difficult
because the influence of a complicated matrix greatly
affects the analytical results. Therefore, different chemical
modifiers are used for the stabilization of the analyte.
In this study, an analytical method was developed for
the determination of Pb, Cd, Fe, Cu, Mn and Zn, in fish
samples, using graphite furnace atomic absorption
spectrometry.
2. Materials and methods
2.1. Apparatus
A Varian Model Spectr AA 220 atomic absorption
spectrometer, equipped with a Varian GTA-110 gra-
phite furnace, was used. Pyrolytic-coated graphite tubes
with a platform were used and signals were measured as
peak areas. The instrument settings and furnace pro-
grammes for analysis of trace elements are described in
Table 1.
2.2. Reagents
were soaked in nitric acid for 15 min and rinsed with
deionized water before use. The stock solutions of metals
(1000 mg/l) were obtained by dissolving appropriate salts
of the corresponding metals (E. Merck) and further dilu-
ted prior to use. High purity Argon was used as inert gas.
2.3. Sampling
The fish species were collected from random com-
mercial catches in the Middle Black Sea coasts in Sam-
sun (Turkey) between September 2000 and May 2001.
The samples were washed with distilled water, dried in
filter paper, homogenized, packed in polyethylene bags
and stored below 20  C until analysis. Five species of
fish, namely Alosa caspia, Engraulis encrasicholus, Tra-
churus trachurus, Sarda sarda, and Clupea sprattus were
included in the study.
2.4. Digestion procedures
Two types of digestion procedures were applied.
Optimum digestion conditions are given below. The
samples were dried to constant weight at 110  C.
2.4.1. Dry-ashing
A sample (1 g) was placed in a high form porcelain
crucible. The furnace temperature was slowly increased
from room temperature to 450  C in 1 h. The samples
were ashed for about 4 h until a white or grey ash resi-
due was obtained. The residue was dissolved in 5 ml of
HNO3 (25% v/v) and the mixture, where necessary, was
heated slowly to dissolve the residue. The solution was
transferred to a 25 ml volumetric flask and made up to
volume (Vaidya & Rantala, 1996). A blank digest was
carried out in the same way. All metals were determined
against aqueous standards.

All reagents were of analytical reagent grade unless 2.4.2. Wet-ashing

otherwise stated. Double distilled water was used for the The samples were solubilized using high-pressure
preparation of solutions. All the plastic and glassware decomposition vessels, commonly known as a digestion

Table 1
Instrument settings and furnace programmes for analysis of trace elements by AAS

Working conditions Fe Cu Mn Zn Pb Cd

Wavelength (nm) 248.3 324.8 279.5 307.6 283.3 228.8

Slit width (nm) 0.2 0.5 0.2 0.7 0.5 0.7
Lamp current (mA) 7 4 5 5 5 4
Ar flow (ml/min) 250 250 250 250 250 250
(flow interrupted at atomization stage)
Injection volume (ml) 20 20 20 20 15 15
Heating programme temperature  C [ramp time (s), hold time (s)]
Drying 1 110(1, 20) 110(1, 20) 110(1, 20) 110(1, 20) 110(1, 20) 110(1, 20)
Drying 2 130(5, 30) 130(5, 30) 130(5, 30) 130(5,30) 130(5,30) 130(5,30)
Pyrolysis 1100(15,10) 1000(15,10) 1300(15,10) 1000(15,10) 900(15,10) 800(15,10)
Atomization 2200(0,5) 2100(0,5) 1900(0,5) 1800(0,5) 2000(0,5) 1900(0,5)
Cleaning 2600(1,2) 2600(1,2) 2400(1,2) 2400(1,2) 2500(1,2) 2400(1,2)
 M. Tüzen / Food Chemistry 80 (2003) 119–123
bomb. A sample (1 g) was placed into Teflon container
and 5 ml of concentrated HNO3 was added. The system
was heated to 130  C for 90 min and finally diluted to 25
ml with deionized water. The sample solution was clear.
A blank digest was carried out in the same way. All
metals were determined against aqueous standards.
2.5. Analytical procedure
Determinations of all metal concentrations were car-
ried out by graphite furnace AAS. During analyses,
internal argon flow rate through the graphite tube was
250 ml/min; gas flow was interrupted during atomiza-
tion. Sample volume, ramp and hold times for the dry-
ing, pyrolysis, atomization and cleaning temperatures
were optimized before analysis to obtain maximum
absorbance and minimum background. The details are
given in Table 1.
Matrix modifiers added were 50 mg (NH4)2HPO4+3
mg Mg(NO3)2 for Cd and Pb, 5 mg Pd+3 mg Mg(NO3)2
for Mn, Fe, Cu, and 5 mg Al2SO4 for Zn.
In order to validate the method for accuracy and
precision, certified reference materials (Oyster tissue and
Mussel tissue) were analysed for each element. The
results are shown in Table 2.
The detection limit is defined as the concentration
corresponding to 3 times the standard deviation of 10
blanks.
3. Results and discussion
The concentrations of the heavy metals in fish species
are given in Table 3. When the dry ashing method is
compared with wet-ashing method for testing metal
concentrations, the t-test indicates that the differences
are not significant at the 95% confidence level, but the
standard deviation of the dry-ashing method is higher
than that of the wet-ashing method. In addition, the
recovery of heavy metals in the dry-ashing method is
lower than that of the wet-ashing method. Dry-ashing is
slow and time-consuming. Wet-ashing favours organic
Table 2
Observed and certified values of elemental concentrations, as mg/g dry weight, in standard reference materials (SRM)
Element NIES-6
Certified value Observed value
Cu 4.90.3 4.8 0.8
Fe 1588 160 12
Mn 16.31.2 15.8 1.9
Zn 1066 105 8
Cd 0.820.03 0.78 0.05
Pb 0.910.04 0.88 0.11
Each value is the average of five determinations. NIES-6: National Institute of Environmental Science, Japan; mussel tissue homogenate. NBS-
1566: National Bureau of Standards and Technology; oyster tissue.
121
matter destruction, shortens the time needed for the
analysis and offers the advantage of simple, fast organic
matter destruction, minimum reagent volume, reduction
of possible analyte losses by volatilization or retention
and elimination of the environmental contamination
risks (Silvestre, Lagarda, Farre, Martinez-Costa, &
Brines, 2000). An open beaker method for dry-ashing
fish samples was found to be susceptible to contamina-
tion from external sources which could be prevented by
carrying out the digestion in the closed system used in
this study.
Detection limits, precision, and accuracy of analyses
were determined by repeated analyses of two biological
reference material standards. The results from the ana-
lysis of SRM were all within the 95% confidence limit of
the SRMs. Excellent recoveries for all metals were
obtained compared to the certified values. The detection
limits for the methods were found to be (mg/l), Cu:0.36,
Mn:0.23, Zn:0.25, Fe:0.42, Pb:0.98 and Cd:0.065 (n:10,
3s). The detection limits of the elements determined in
this study were found to be 10 times and 100 times
lower than those reported in other studies using flame
AAS (Karadede & Ünlü, 2000; Kress et al., 1998).
It is desirable to use a higher pyrolysis temperature in
order to remove the matrix efficiently and a temperature
of at least 1000  C should be aimed at for many analytes
in food, biological and environmental samples (Bin &
Zhe-Ming, 1996; Lynch & Littlejohn, 1989; Zong, Par-
sons, & Slavin, 1996). The pyrolysis and atomization
temperatures of heavy metals were increased using dif-
ferent chemical modifiers. The maximum pyrolysis and
atomization temperatures obtained were 800 and
1900  C for Cd; 900 and 2000  C for Pb; 1000 and
1800  C for Zn; 1300 and 1900  C for Mn; 1000 and
2100  C for Cu; 1100 and 2200  C for Fe. Most of the
matrix was removed before the atomization step and
less interference occurred during atomization. The
method of standard additions was used to avoid the
matrix effect.
The fish species analysed are used for human con-
sumption. All metal concentrations were determined on
a dry weight basis. According to the results (Table 3),
Recovery (%) NBS-1566 Recovery (%)
Certified value Observed value
98 63.0 3.5 64.42.8 102
101 195 34 20010 103
97 17.5 1.2 17.91.8 102
99 852 14 86110 101
95 3.50 0.40 3.550.65 101
97 0.48 0.04 0.500.10 104
122 M. Tüzen / Food Chemistry 80 (2003) 119–123

Table 3
Metal concentrations, as mg/g dry weight, in fish species (the number of determinations on each sample is 10)

Sample No.a Methodb Cd Pb Cu Fe Mn Zn

1 A 0.350.05 0.52 0.16 2.930.18 16.081.15 1.57 0.24 20.411.75

B 0.340.08 0.51 0.21 2.900.31 15.502.10 1.56 0.14 22.941.60

2 A 0.200.03 0.38 0.02 1.940.10 10.451.63 1.96 0.12 17.382.01

B 0.180.02 0.39 0.07 1.960.17 10.321.05 1.98 0.32 18.851.72

3 A 0.470.10 0.85 0.16 1.520.35 32.402.70 3.76 0.45 12.052.30

B 0.480.08 0.83 0.36 1.550.26 31.261.73 3.50 0.58 11.411.15

4 A 0.090.02 0.22 0.04 1.280.14 9.520.81 1.06 0.27 11.201.44

B 0.100.01 0.26 0.07 1.290.32 10.141.11 1.33 0.42 13.721.32

5 A 0.300.15 0.74 0.11 1.790.62 25.483.18 2.82 0.24 9.500.60

B 0.300.28 0.68 0.17 1.830.44 24.122.06 2.74 0.44 10.361.29
a
1, Alosa caspia; 2, Engraulis encrasicholus; 3, Trachurus trachurus; 4, Sarda sarda; 5, Clupea sprattus.
b
a, Dry ash; b, wet ash.

the metal contents in the samples studied depend on the
analyzed species. The concentration of cadmium varied
from 0.09 to 0.48 mg/g; for lead it ranged from 0.22 to
0.85 mg/g. The lowest and highest values of these ele-
ments were found in Sarda sarda and Trachurus tra-
churus species, respectively. Levels of the essential
metals in the fish samples were higher than those of the
non-essential metals. Among the six metals under study,
zinc showed the highest level of accumulation. A similar
situation was observed in studies (Cid et al., 2001). The
concentration of zinc varied from 9.50 to 22.94 mg/g.
The lowest and highest values of this element were
observed in C. sprattus and A. caspia species, respec-
tively. The concentrations of copper, iron and manga-
nese in the S. sarda species were lower than those found
in other analysed fish species. The highest iron and
manganese values were found in T. trachurus species.
The fact that toxic metals are present in high con-
centrations in marine organisms is of particular impor-
tance in relation to the FAO/WHO (1976) standards for
Pb and Cd as toxic metals. The maximum permissible
doses for an adult are 3 mg Pb and 0.5 mg Cd per week,
but the recommended doses are only one-fifth of those
quantities.
Results achieved for heavy metals in fish samples
collected from the middle Black Sea, Turkey were in
good agreement with other reported data from the lit-
erature.
4. Conclusions
The proposed method was efficient for simple, rapid
and reliable determination of some heavy metals in the
fish species. The accuracy of the method was checked
and confirmed by standard reference materials. The
recoveries of the heavy metals in standard reference
materials were in the range of 95–104%. The relative
standard deviations were less than 7%. The wet-diges-
tion in closed vessels was the best. It was suitable for
routine analysis. It was found that the concentrations of
heavy metals in the fish samples were below those of
Public Health Regulation in Turkey (Anonymous,
1995).
References
Acar, O. (2001). Determination of cadmium and lead in biological
samples by zeeman ETAAS using various chemical modifiers.
Talanta, 55(3), 613–622.
Acar, O., Kılıç, Z., & Türker, A. R. (2000). Determination of lead in
cookies by electrothermal atomic absorption spectrometry with
various chemical modifiers. Food Chemistry, 71, 117–122.
Anonymous. (1995). Official Gazzette, number 22223, p. 1361 (in
Turkish).
Balkas, I. T., Tugrel, S., & Salhogln, I. (1982). Trace metal levels in
fish and crustacea from Northeastern Mediterranean coastal waters.
Marine Environmental Research, 6, 281–289.
Bin, H., & Zhe-Ming, N. (1996). Minimization of sulfate interference
on lead atomization with palladium-strontium nitrate as chemical
modifier in electrothermal atomic absorption spectrometry. Journal
of Analytical Atomic Spectrometry, 11, 165–168.
Blust, R., Vander der Linden, A., Verheyen, E., & Decleir, W. (1988).
Evaluation of microwave heating digestion and graphite furnace
atomic absorption spectrometry with continuum source background
correction for the determination of iron, copper and cadmium in
brine shrimp. Journal of Analytical Atomic Spectrometry, 3, 387–393.
Cid, B. P., Boia, C., Pombo, L., & Rebelo, E. (2001). Determination of
trace metals in fish species of the Ria de Aveiro (Portugal) by electro-
thermal atomic absorption spectrometry. Food Chemistry, 75, 93–100.
Doner, G., & Akman, S. (2000). A comparision of sample preparation
procedures for the determination of iron and zinc in bulgur wheat
by graphite furnace atomic absorption spectrometry. Analytical let-
ters, 33(15), 3333–3341.
FAO/WHO. (1976). List of maximum levels recommended for con-
taminants by the Joint FAO/ WHO Codex Alimentarius Commis-
sion. Second Series. CAC/FAL, Rome, 3, 1–8.
Giordano, R., Arata, P., Ciaralli, L., Rinaldi, S., Giani, M., Cicero,
 M. Tüzen / Food Chemistry 80 (2003) 119–123
A. M., & Costantini, S. (1991). Heavy metals in mussels and fish
from Italian coastal waters. Marine Pollution Bulletin, 22(1), 10–14.
Huang, Y. L., Chuang, I. C., Pan, C. H., Hsiech, C., Shi, T. S., & Lin,
T. H. (2000). Determination of chromium in whole blood and urine
by graphite furnace AAS. Atomic Spectroscopy, 21(1), 10–16.
Karadede, H., & Ünlü, E. (2000). Concentrations of some heavy
metals in water, sediment and fish species from the Atatürk Dam
Lake (Euphrates), Turkey. Chemosphere, 41, 1371–1376.
Knapp, G. (1991). Mechanized techniques for sample decomposition
and element preconcentration. Microchimica Acta, 2, 445–455.
Kress, N., Hornung, H., & Herut, B. (1998). Concentrations of Hg,
Cd, Cu, Zn, Fe and Mn in deep sea benthic fauna from the South-
eastern Mediterranean Sea: a comparison study between fauna col-
lected at a Pristine area and at two waste disposal sites. Marine
Pollution Bulletin, 36(11), 911–921.
Lynch, S., & Littlejohn, D. (1989). Palladium as a chemical modifier
for the determination of lead in food slurries by electrothermal ato-
misation atomic absorption spectrometry. Journal of Analytical
Atomic Spectrometry, 4, 157–161.
Pujin, V., Djukic, N., Maletin, S., Obradovic, S., & Kostic, D. (1990).
Content of heavy metals in some fish species in the section of the
Danube flowing through Vojvodina. Water Science & Technology,
22(5), 79–86.
Santos, H. C. D., Korn, M. G. A., & Ferreira, S. L. C. (2001).
Enrichment and determination of molybdenum in geological sam-
ples and seawater by ICP-AES using calmagite and activated car-
bon. Analytica Chimica Acta, 426, 79–84.
Schroeder, H. A. (1973). The trace elements and nutrition. London:
Faber and Faber.
Sharif, A. K. M., Mustafa, A. L., Mirza, A. H., & Safiullah, S. (1991).
Trace metals in tropical marine fish from the Bay of Bengal. Science
of the Total Environment, 107, 135–142.
Silvestre, M. D., Lagarda, M. J., Farre, R., Martinez-Costa, C., &
123
Brines, J. (2000). Copper, iron and zinc determinations in human milk
using FAAS with microwave digestion. Food Chemistry, 68, 95–99.
Somer, E. (1974). Toxic potential of trace metals in foods. A review.
Journal of Food Science, 39, 215–217.
Sures, B., Taraschewski, H., & Haug, C. (1995). Determination of
trace metals (Cd, Pb) in fish by electrothermal atomic absorption
spectrometry after microwave digestion. Analytica Chimica Acta,
311(2), 135–139.
Tarýq, J., Jaffar, M., & Ashraf, M. (1991). Levels of selected heavy
metals in commercial fish from five freshwater lakes, Pakistan.
Toxicological and Environmental Chemistry, 33, 133–140.
Tarýq, J., Jaffar, M., & Moazzam, M. (1991). Concentration corre-
lations between major cations and heavy metals in fish from the
Arabian Sea. Marine Pollution Bulletin, 22(11), 562–565.
Ubillus, F., Alegria, A., Barbera, R., Farre, R., & Lagerda, M. J.
(2000). Methylmercury and inorganic mercury determination in fish
by cold vapour generation atomic absorption spectrometry. Food
Chemistry, 71, 529–533.
Vaidya, O. C., & Rantala, R. T. T. (1996). A comparative study of
analytical methods: determination of heavy metals in mussels
(Mytilus edulis) from Eastern Canada. International Journal of
Environmental & Analytical Chemistry, 63, 179–185.
Voegborlo, R. B., El-Methnani, A. M., & Abedin, M. Z. (1999).
Mercury, cadmium and lead content of canned tuna fish. Food
Chemistry, 67, 341–345.
Zhuang, Z. X., Wang, X. O., Yang, C. L., Zhu, E., & Yang, P. Y.
(1995). Chemometric analysis of coastal water pollution by exami-
nation of marine organisms using atomic spectrometry. Micro-
chemical Journal, 51(1), 138–144.
Zong, Y. Y., Parsons, P. J., & Slavin, W. (1996). Accurate and precise
measurements of lead in bone using electrothermal atomic absorp-
tion spectrometry with zeeman-effect background correction. Jour-
nal of Analytical Atomic Spectrometry, 11, 25–30.